DataTrain.AI

LLaVA-OneVision-2-Data Training data for the LLaVA-OneVision-2 multimodal model family, covering large-scale video and spatial reasoning corpora used…

<1K·Apache-2.0·Parquet

Image Multimodal Video

Goku Seedance 2 Video Prompts Dataset

1K–10K·CC-BY

Image

COCO

330K images with object, segmentation and caption labels.

100K–1M·CC-BY·COCO JSON

Image

ImageNet

14M+ images across 20K+ categories (1K subset common).

>10M·Custom / Research-only·Images (folder)

E

Encord

Labeling & Annotation

Encord is a multimodal data platform for annotating and curating images, video, DICOM, text and 3D/LiDAR data, with model evaluation…

3D / Point Cloud Image LiDAR / Radar Video

V

V7

Labeling & Annotation

V7 builds an AI data engine spanning V7 Darwin for vision training data and V7 Go for document and knowledge-work…

Image Video

R

Roboflow

Labeling & Annotation

Roboflow is a computer-vision developer platform for labeling, training and deploying models, plus Universe, a large open library of datasets…

Image Video

C

CVAT.ai

Labeling & Annotation

CVAT.ai maintains CVAT, the open-source Computer Vision Annotation Tool for image, video and 3D labeling, offered as self-hosted, cloud and…

Image Video

V

Voxel51

Data Processing & Curation

Voxel51 develops FiftyOne, an open-source toolkit for curating, visualizing and analyzing visual AI datasets, helping teams find and fix data…

Image Video

F

FiftyOne

Exploration & Visualization

Open source

Open-source dataset curation and visualization toolkit by Voxel51 for exploring, debugging, and refining visual AI datasets and model results in…

Image Video

Voxel51Freemium

R

Roboflow Annotate

Annotation & Labeling

Web-based computer vision annotation and dataset management from Roboflow, with AI-assisted auto-labeling, smart polygon tools, and dataset versioning.

Image Video

RoboflowFreemium

E

Encord Annotate

Annotation & Labeling

Encord's annotation platform for image, video, 3D/LiDAR, DICOM, and SAR data, with SAM-based automation and human-in-the-loop review for complex domains.

3D / Point Cloud Image LiDAR / Radar Video

EncordEnterprise

V

V7 Darwin

Annotation & Labeling

V7 Darwin is an image and video annotation platform with SAM-based auto-annotation, video auto-tracking, and configurable multi-stage review workflows.

Image Video

V7Paid

Image Multimodal Text

LAION-5B

5.85B CLIP-filtered image–text pairs.

>10M·CC-BY·Parquet

N

Nimble

Web / Real-time Data

Real-time web data platform using AI-powered web agents to turn public pages into structured, validated data for AI applications, agents…

Image Text

O

Oxylabs

Web / Real-time Data

Proxy network and web-scraping API provider delivering large-scale public web data collection infrastructure for AI, market research and analytics teams.

Image Text